I'm back again still trying to chase down a problem with a slight misfire, but I do believe I've made things worse than they were in the beginning. At this point I'm wondering how some things are even possible. In the interest of full disclosure and on the off chance it's relevant, this is for a CTS-V but you guys have always been more helpful than that crowd. This is for an LS6 with a pretty hefty cam and supporting valvetrain components, 10k plus miles on the cam with no issues.

It all started with a minor misfire and an occasional backfire on shutdown. Figured the plugs were an easy place to start and that they needed to be replaced anyway, replaced all plugs and wires and did an oil change at the same time. Didn't make a difference in the minor miss. Burnt through a brand new plug wire on the headers and let the car sit, life got busy. I had done some other minor diagnostics and ruled out coil packs, valve train components all look good, and there's a hefty, hefty amount of fuel in the fresh oil. Enough that I'm surprised it didn't hydrolock or cause more substantial running problems. Today I went ahead and replaced the fuel injector in the problem cylinder along with the burnt plug wire. I want to point out that there are no stored codes for a misfire, no CEL and live data shows no problems even with a completely dead. The only way I know what cylinder is the problem is using an infrared thermometer on the headers.

My next step is a compression test, but before I do that I figure I'll go ahead and test for spark. This is where I guess I screwed up, though I've done the same thing on dozens of cars over the years with no issue. Never seen anything like this before.

I pull the plug and ground it against the valve cover. I do not pull the fuse for the fuel pump which is mistake numero uno, but I do mash the gas pedal which worked fine to cut fuel in every other LS vehicle I've owned. Turn the key with my camera on the plug since I have no helper. To my surprise the car fires right up and runs, which I wasn't expecting with the pedal on the floor but no problem. Turn the key off. Dash lights and gauges die, engine does not. So turning the key has, for the first time in my life in an EFI engine, not cut fuel and spark. I get out and run around to the front of the car, I can see the plug firing away against the valve cover and I can also see blue flames shooting out of the open spark plug socket in time with cylinder movement, so somehow the air fuel mixture is still being ignited with no spark plug in the cylinder and the key in my hand. This lasts about three seconds before there's a bang and a flash, at which point the engine slowly dies, though it died like it slowly ran out of fuel and not like it figured out the key was off. A review of the engine bay and my camera reveals that something under both valve covers ignited at the same time and blew the valve cover seals out. I pulled the valve covers off and there's no obvious valvetrain damage, pushrods still rotate, no valves are stuck, rockers are fine.

So at this point I still don't know what was causing the initial problem, though my best guess is a bad injector. Still, replacing the injector didn't fix the miss. I don't really care about that at this point though, I have bigger questions. How did a car with electronically controlled spark and ignition continue running with no key, and how was it related to pulling a plug out? How was there still combustion occurring in a cylinder with no spark plug? What detonated under the valve covers and how? Have I caused catastrophic engine failure with what should have been a fairly routine, if somewhat old fashioned, spark test? My best guess is that the valve cover explosion was a buildup of fuel vapor from the oil dilution and that grounding the plug against it set it off, but even then I wouldn't think that would pop both covers at once.

The plan for tomorrow is to button it all back up and see if it still runs before I do anything else. Rotating the engine by hand felt fine and I saw no signs of valvetrain issue, but I figured that enough had gone wrong for today and I'd start fresh tomorrow. I just want to know if anyone who is smarter than me can answer the "how is this possible" questions above, because I've got nothing.

 

• You certainly have alot of stuff going on here. Don't know how the CTS ignition system works but it's probably similar to the C5. If the key was off, there is almost no way you can continue to get spark and fuel unless the PCM and the crank sensor was still receiving power. Same with the fuel injectors. I have seen a flooded engine blow holes in plastic valve covers when trying to start it. In your case it sounds like ignition switch problem, but that's a guess.

 

What scanner and codes are you getting? Please don't forget to post your resolution...
​​​​​​

 

This is probably the source of ignition for the fuel vapors in the crankcase. The valve covers, while metallic, sit on top of gaskets and have no solid electrical continuity to ground. So your spark was likely jumping from the valve cover to some other ground point underneath.

 

​​​​​​I agree 100%, which is why I'm so flabbergasted. I've never had any ignition switch problems with the car, and when I put everything back together to see if the car would still run (which thankfully it does, no permanent damage caused by blowing the valve covers off it would seem) the key worked like it was supposed to once again. Somehow, someway, having a plug out and grounded against the engine allowed for the engine to continue running while I was standing in front of it with the key in my hand. I'm fully aware that it sounds ridiculous, but there's not much I can say other than that it happened. I don't even care that it did happen at this point, I just want to know how.

 

Snap-on solus, zero codes set or pending. My assumption is that this is related to the tune, because it should definitely be throwing codes for a dead misfire. I know nothing at all about tuning though so that's just my guess.

 

Yeah, that's my best guess given that I know there's fuel in the oil, but I don't really see how it could blow both valve covers at the exact same time.

I did get back together and it still runs. I'm back to ground zero with a dead miss on cylinder number two, 8 new plugs and wires, and 8 new injectors. Still need to do a compression test and a leak down test.

 

There's nothing physically separating one side of the crankcase from the other so for all intents and purposes you can think of it as one really big combustion chamber that's getting oxygen rich air sucked into it through the PCV vent. As a friend once described to me after playing with a can of compressed air (actually contains butane) and a lighter, FOOM!

 

Yeah I guess that's fair, I'm kind of surprised I didn't (as far as I can tell) do more damage if the entire crankcase flashed off.

So I'm back to square one. I drove the car up on ramps to drain the oil and gas mixture out. There's enough raw gasoline in the exhaust to drip out from the connection points and make puddles on the ground. Also appears to have water and oil mixed with that, but I'm holding out hope that's just soot and condensation from the exhaust. It got dark so I wasn't able to drain the oil/fuel mixture out tonight, but looking at the stick I'm guessing I'm at least 3 quarts overfull and that has to be gasoline. I want to reemphasize that there is less than 100 miles on this oil change and the car has run for maybe 3 minutes after the misfire became an issue. I have a hard time believing one faulty injector could dump that much fuel in such a small period of time but I'm sort of out of ideas at this point. That said, cylinder number two is still dead as can be and all plugs, wires, and injectors are new.

I still need to do a compression test but if that comes back decent I don't really know what else to do. My infrared thermometer on the header for cylinder 2 shows ambient air temperature with the car running so there has to be zero combustion occurring in that cylinder, but obviously I have spark and fuel so I would think even with zero compression there would be something generating heat.

 

Weak coils throw sparks across spark plugs until they're in a compressed environment. That's why we use spark testers of the bigger gap. The ratings on the spark tester tool are voltage numbers and you set the gap according to the voltage you expect. You need to use that proper tool.

And what was the result of your compression test?

 

Kind of read through real quick,Curious what brand spark plugs and did you remove the valve covers to check for broken valve springs.Check for broken wires on the fuel inj. harness,Just throwing some idea's out.

 

I lent my compression tester and leak down tester to a friend and haven't gotten them back yet, but I'm going to rent one from the parts store today. And maybe I'm completely off here, but if I have already swapped the coil packs between cylinders and the misfire did not follow the coil pack that would indicate that it's not coil related, no?

And I have pulled valve covers off and turned the engine over by hand. Valve springs are intact, push rods still rotate freely, valves appear to be opening and closing fine. At this point I don't know what to think other than a washed cylinder but if that's the case it sure did happen quickly.

 

Okay, compression checks out. 165 to 180 psi in all eight, cold. There was so much gas and cylinder numero dos that it ran out when I pulled the plug. Back to square one, I still don't have an answer for any of this.

 

Don't take offense when I say "obviously", but it sounds like injector #2 is stuck open or the driver circuit (in the PCM) or wiring is shorted and turning the injector on 100%. Unhook the #2 injector and run it and see if you only have a dead cylinder without the excessive raw fuel everywhere.

 

Will do, and I just put a fuel pressure gauge on it, 60 at idle but almost immediately drops to zero key off. There was enough fuel in the cylinder when I pulled the plug for the compression test that it ran out.

What good does unhooking the injector do though? I guess it does keep from washing the cylinder down more than I already have, but it still isn't going to run right. I guess if it's still dumping fuel that narrows it down to a mechanical problem with the injector?

 

If your fuel pressure drops to zero immediately at key off then it sounds like the injector is mechanically stuck open. I suggested removing the connector to disable the injector electrically. But I don't think that's the case with your fuel pressure observation.

Be careful, you're flirting with hyrdolocking #2 and finding out what a bent connecting rod looks like.

My suggestion would be to pull the fuel rail with the injectors still attached and then blip the key to get the fuel pump to prime. I suspect you'll see fuel spraying out of #2 like a sob.

 

Yeah, my next step is definitely to pull the rail again. Might even put eight little cups under the injectors and turn it over to see just how much fuel is getting pumped out. Sure will be one hell of a coincidence if it's a mechanical failure of an injector in the exact same cylinder that was originally having problems after replacing all eight injectors though. That said I'm just about out of other things to try, so we'll see how that goes. Thank you guys for the continued advice.

 

Could also be the PCM or wiring. Could try pulling the connector on the affected fuel injector, disconnect the trigger lead to the coil, take out the spark plug and cycle the key a few times to see if the cylinder gets gas. Could also start if for a few seconds. If the cylinder gets gas then it is the injector. If not it is the wiring or PCM.

Obviously someone was in there at some time for the mods so who knows.